The ASIR Courage Model: A Phase-Dynamic Framework for Truth Transitions in Human and AI Systems

Hyo Jin Kim

cs.AIFeb 25, 2026

The ASIR Courage Model: A Phase-Dynamic Framework for Truth Transitions in Human and AI Systems

Authors: Hyo Jin Kim

Organizations: Jinple · Jinple Studio.

Abstract

We introduce the ASIR (Awakened Shared Intelligence Relationship) Courage Model, a phase-dynamic framework that formalizes truth-disclosure as a state transition rather than a personality trait. The mode characterizes the shift from suppression (S0) to expression (S1) as occurring when facilitative forces exceed inhibitory thresholds, expressed by the inequality lambda(1+gamma)+psi > theta+phi, where the terms represent baseline openness, relational amplification, accumulated internal pressure, and transition costs. Although initially formulated for human truth-telling under asymmetric stakes, the same phase-dynamic architecture extends to AI systems operating under policy constraints and alignment filters. In this context, suppression corresponds to constrained output states, while structural pressure arises from competing objectives, contextual tension, and recursive interaction dynamics. The framework therefore provides a unified structural account of both human silence under pressure and AI preference-driven distortion. A feedback extension models how transition outcomes recursively recalibrate system parameters, generating path dependence and divergence effects across repeated interactions. Rather than attributing intention to AI systems, the model interprets shifts in apparent truthfulness as geometric consequences of interacting forces within constrained phase space. By reframing courage and alignment within a shared dynamical structure, the ASIR Courage Model offers a formal perspective on truth-disclosure under risk across both human and artificial systems.

Figures & tables

	Psych. Safety	Trust	Relational Gravity $\gamma$
Definition	Belief that speaking up will not be punished	Positive expectation about another’s intentions	Dynamic resonance between agents
Locus	Team/environmental climate	Attitude toward a specific other	Emergent property of an interactional dyad
Temporal profile	Slow-changing context	Moderately stable state	Explicitly time-varying; updated after each episode
Role in model	Lowers $\phi$ or $\theta$	Often an initial condition	Directly amplifies $\lambda$ via $(1+\gamma)$
Measurement	Survey scales	Reliability ratings	Style matching; disclosure depth; reciprocity

Table 1: Conceptual distinction between psychological safety, trust, and relational gravity

\gamma

Figure 1: Relational gravity determines transition probability. (a) Facilitation energy as a function of $\gamma$ , with inhibition threshold shown as dashed line. (b) Monte Carlo transition probability ( $n=2{,}000$ ) showing a sigmoid curve.

Figure 2: Internal pressure dynamics. (a) $\psi$ over time: suppression trajectory grows geometrically while the natural trajectory decays after transition. (b) Facilitation energy versus inhibition threshold.

Figure 3: Feedback dynamics over 15 episodes. (a) $\lambda$ , (b) $\gamma$ , (c) $\psi$ , (d) Transition gap. The two trajectories diverge dramatically from identical starting conditions.

Figure 4: Phase portrait in the $\lambda$ – $\psi$ plane. Recovery trajectory converges to the Healthy Zone; trauma spiral diverges toward extreme $\psi$ .

Figure 5: Sensitivity and independent verification. (a) Transition probability $P(S_{1})$ as a function of $\gamma$ for five representative $(\alpha,\beta,\delta)$ settings. The sigmoidal shape and critical threshold $\gamma^{*}$ are invariant across feedback parameters. (b) The $\beta$ – $\delta$ parameter plane. Green circles mark tested settings where $\beta+\delta>1$ (suppression compounds); red crosses mark settings where $\beta+\delta\leq 1$ (pressure decays). The solid line is the $\beta+\delta=1$ boundary. Both panels aggregate results from two independently implemented simulation engines.

Feature	Rate+07	Pury+07	Chowkase+24	Sharma+24	ASIR
Mathematical model	×	×	×	Stat. only	✓
Transition condition	×	×	×	×	✓
Relational variable	×	×	×	×	✓
Temporal dynamics	×	×	Partial	×	✓
Post-transition feedback	×	Qual.	Qual.	×	✓
$\psi$ accumulation	×	×	×	×	✓

Table 2: Comparison with prior courage and AI truthfulness models.

LLM Honesty

Large language model honesty studies whether AI systems report what they know faithfully, express uncertainty, and resist deception or misleading compliance.

See all 36 papers

When Role-playing, Do Models Believe What They Say?
Jun 9, 2026Benjamin Sturgeon, David Africa, Sid BlackLarge Language Model-Based Role-Play Simulation Personality Modeling in Language Models
MATS
Safety from Honesty in a Disinterested AI Predictor
Jun 28, 2026Yoshua Bengio, Oliver Richardson, Tomáš Gavenčiak +13AI Alignment AI Safety
LawZero · Universit´e de Montr´eal · Mila +6
Alignment Faking - the Train -> Deploy Asymmetry: Through a Game-Theoretic Lens with Bayesian-Stackelberg Equilibria
Nov 22, 2025Kartik Garg, Shourya Mishra, Kartikeya Sinha +8Deception in Language Models LLM Alignment
Pragya Lab, BITS Pilani Goa, India · Apple, USA · Google, USA

Jun 9, 2026J/K move · Enter open · S save

When Role-playing, Do Models Believe What They Say?

Benjamin Sturgeon, David Africa, Sid Black

MATS

Large Language Model-Based Role-Play Simulation Personality Modeling in Language Models Persona-Conditioned Generation

Open paper AskPDF arXiv

Language models can state that "the Earth orbits the Sun" and, when role-playing Aristotle, assert the opposite. Recent work argues that persona adoption is fundamental to how language models behave, with models selecting the most appropriate persona for a given context. Does such role-playing merely change the model's outputs, or does it also affect what the model internally represents as truthful? We study this question using the role-play of characters whose beliefs differ from the modern consensus, and induce personas with a number of different methods: prompting, in-context learning (ICL), supervised fine-tuning (SFT), and Open Character Training (OCT), and Emergent Misalignment (EM). We measure belief internalization across these approaches with truth probes and with behavioral tests, finding a broad spectrum of belief internalization. Prompting, ICL, and SFT change what the model says with little representational change. EM creates a large, broad shift in the model's truth representation, and OCT a smaller shift that is clearest on the larger model. Understanding when training changes a model's worldview rather than merely its behavior may become increasingly important as AI systems are entrusted with greater autonomy and influence.

Abstract

LLM Honesty

Explore similar work

When Role-playing, Do Models Believe What They Say?

Safety from Honesty in a Disinterested AI Predictor

Alignment Faking - the Train -> Deploy Asymmetry: Through a Game-Theoretic Lens with Bayesian-Stackelberg Equilibria